Читать книгу Dry Beans and Pulses Production, Processing, and Nutrition - Группа авторов - Страница 42

For a complete description of approaches used to improve yield in beans, the reader is referred to the review by Kelly (2018). Many genetic studies in the literature focus on specific traits by necessity as they present important information for breeders on genetic inheritance, heritability, location, and effect of QTL. However, bean breeders must combine multiple economic traits into a single successful variety with a combination of favorable traits with few minor deficiencies. When selection is practiced for individual traits, invariably other traits are affected, and unless selection is practiced to maintain them at optimum levels, they will tend to diminish. In the 1970s, the breeding program at Michigan broadened the genetic base of navy beans by crossing with black beans from Central America (Adams 1982). The resulting progeny were higher yielding with favorable upright architecture, but they lacked satisfactory canning quality present in the older navy bean varieties (Hosfield and Uebersax 1980). Correcting this problem required another series of crosses to reintroduce favorable canning quality traits. In order to maintain stability in most traits, a three‐tiered breeding pyramid (Figure 2.5, Kelly et al. 1998) based on the level of genetic variability available to the breeder was proposed as the most effective and flexible method to enhance yield in beans. Different breeding methods were proposed, depending on whether a private breeder was only crossing elite lines from the same market class, in contrast to a public sector USDA breeder working to enhance germplasm using wide genetic crosses, whereby the goal might not be an actual commercial variety.

Fig. 2.5. Breeding pyramid. A three‐tiered approach to breeding for yield in common bean.

Source: Kelly et al. (1998).

The choice of breeding systems is dictated by contrasting goals as well as the type of germplasm and the traits being improved. Having flexibility to change methods is critical due to the unpredictability of genetic recombination in certain crosses. Over the years, many breeders have identified specific genotypes/varieties not only as specific donor of economic traits but as good general combiners (i.e., they generally produce useful progeny). Other varieties that may be high yielding may prove to be poor general combiners, producing an array of mediocre progeny regardless of the other parents used in the cross. This information is only accumulated after years of experience of making crosses and from sharing similar information with breeder colleagues.